I am a researcher in computer science, specializing in databases and human-AI interaction. In 2027, I will join Carnegie Mellon University as an Assistant Professor in the Computer Science Department (CSD) and, by courtesy, the HCII.

My research goal is to build the "full stack" (i.e., systems and interfaces) for data-intensive knowledge work. We take an applications-oriented approach: we build open-source software, deploy with real users, and use what we learn to advance research in systems, interfaces, and user behavior.

Current Research Projects

• Efficient & accurate AI-powered data processing
• AI agent observability
• Async interfaces for human-AI collaboration

I (am close to having) received my PhD in EECS from UC Berkeley's Data Systems and Foundations group, advised by Aditya Parameswaran. In my PhD, I designed and built the DocETL and DocWrangler ecosystem for scalable LLM-powered data processing. I also wrote several papers, a course, and a book on evaluating LLM-powered applications. Our work has had real-world impact in (1) databases, e.g., Snowflake, BigQuery; (2) AI tooling, e.g., LangChain, ChromaDB, OpenAI; and (3) society, e.g., DocETL powers a tool helping California public defenders challenge wrongful convictions under the Racial Justice Act.